<p>Short-read genome projects in non-model fishes are often interpreted in chromosome-scale coordinates before assemblies can support chromosome-scale claims. We examined this problem in the Qiantang River <i>Megalobrama</i> sample SJF by projecting one polished Illumina-derived substrate onto two chromosome-level <i>Megalobrama amblycephala</i> references, defined as reference A (National Center for Biotechnology Information) and reference B (Genome Warehouse). Using 670,751,604 filtered reads (100.14&#xa0;Gb), we first characterized genome architecture and compared candidate short-read substrates before projection. The analysis separated substrate quality from projected-coordinate interpretation. K-mer spectra estimated a 906.7–909.0&#xa0;Mb haploid genome, 1.18–1.20% heterozygosity, and 34.7% repeat content, consistent with fragmentation in an Illumina-only draft. The raw MaSuRCA draft spanned 1.072&#xa0;Gb across 519,765 scaffolds, whereas the Platanus donor draft was inflated (1.874&#xa0;Gb). NextPolish refinement preserved draft-level Benchmarking Universal Single-Copy Orthologs (BUSCO) completeness (34.1%) while improving consensus and read-backed support, so the NextPolish-refined MaSuRCA derivative was carried forward. Projection recovered 95.5–97.9% complete BUSCOs as projected gene-space representation. Reference choice affected anchored span much more than the correction step: reference A anchored 919.17–919.24&#xa0;Mb, whereas reference B anchored 878.98–879.06&#xa0;Mb. Yet 50-kb concordance remained high (92.47–96.95% one-to-one), callable fractions were nearly identical (77.24–77.35%), and callable core sequence reached 724.85–727.35&#xa0;Mb. Gene-interval partitioning further placed most transferred gene-bearing sequence in the callable core. The benchmark therefore defines a large callable core and a smaller reference-sensitive shell that delimit where local sequence and reference-addressed interpretation are best supported in a fragmented short-read draft.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Dual-reference projection defines a callable core and reference-sensitive shell in a Qiantang River Megalobrama draft assembly

  • Kai Liu,
  • Wei Zhang,
  • Jialing Qiao,
  • Nan Xie

摘要

Short-read genome projects in non-model fishes are often interpreted in chromosome-scale coordinates before assemblies can support chromosome-scale claims. We examined this problem in the Qiantang River Megalobrama sample SJF by projecting one polished Illumina-derived substrate onto two chromosome-level Megalobrama amblycephala references, defined as reference A (National Center for Biotechnology Information) and reference B (Genome Warehouse). Using 670,751,604 filtered reads (100.14 Gb), we first characterized genome architecture and compared candidate short-read substrates before projection. The analysis separated substrate quality from projected-coordinate interpretation. K-mer spectra estimated a 906.7–909.0 Mb haploid genome, 1.18–1.20% heterozygosity, and 34.7% repeat content, consistent with fragmentation in an Illumina-only draft. The raw MaSuRCA draft spanned 1.072 Gb across 519,765 scaffolds, whereas the Platanus donor draft was inflated (1.874 Gb). NextPolish refinement preserved draft-level Benchmarking Universal Single-Copy Orthologs (BUSCO) completeness (34.1%) while improving consensus and read-backed support, so the NextPolish-refined MaSuRCA derivative was carried forward. Projection recovered 95.5–97.9% complete BUSCOs as projected gene-space representation. Reference choice affected anchored span much more than the correction step: reference A anchored 919.17–919.24 Mb, whereas reference B anchored 878.98–879.06 Mb. Yet 50-kb concordance remained high (92.47–96.95% one-to-one), callable fractions were nearly identical (77.24–77.35%), and callable core sequence reached 724.85–727.35 Mb. Gene-interval partitioning further placed most transferred gene-bearing sequence in the callable core. The benchmark therefore defines a large callable core and a smaller reference-sensitive shell that delimit where local sequence and reference-addressed interpretation are best supported in a fragmented short-read draft.